The primary objective of the current proposal is to develop new criteria which will identify patients in need of adjuvant therapy and differentiate these patients from those in whom surgical therapy alone is curative. Many oncologists rely on primary tumor depth and clinically detectable disease in their determination of applicable therapy. The management of melanoma can be improved if techniques to detect micrometastases are developed to augment those in current practice. During the last grant period we developed innovative lymphatic mapping and selective lymphadenectomy techniques (Project III). These techniques have allowed us to detect occult melanoma cells by intraoperative frozen sections in regional lymph nodes of clinical Stage I melanoma patients. This technique accurately identifies patients with early stage melanoma who have nodal metastases and are likely to benefit from radial lymphadenectomy. Furthermore, it avoids unnecessary lymph node dissection in those patients whose tumors failed to metastasize. This new approach will be extended to the multicenter cooperative study in the current year. On the other hand, a substantial number of Stage I melanoma patients with occult-node metastases do not benefit from lymphadenectomy since they die from their disseminated metastases. Furthermore, 20% of all patients without occult metastases detected by H&E staining and/or immunostaining recur within 5 years despite resection of their nodes. It is obvious that these patients have subclinical systemic metastases at the time of initial surgical treatment. Therefore, we propose to investigate methods to identify Stage I patients who may require adjuvant therapy. Various histological, immunological, and molecular/biochemical methods will be evaluated.